The Worst Ancova Model Among All Possible Models With The Same Set Of Variables

Ngurah Agung, I Gusti and Annas, Suwardi (2015) The Worst Ancova Model Among All Possible Models With The Same Set Of Variables. In: International Conference on Statistics, Mathematics, Teaching, and Research, 9 - 10 Oktober 2015, Makassar, Indonesia.

[img] Text (Artikel Prosiding Internasional)
3. A Worst_Artikel 3.pdf - Published Version
Download (1MB)
[img] Text (Peer Review Artikel Prosiding Internasional)
Peer Review Artikel Prosiding Internasional - The Worst Ancova Model Among All Possible Models With The Same Set Of Variables.pdf - Published Version
Download (2MB)
[img] Text (Turnitin Artikel Prosiding Internasional)
Turnitin Artikel Prosiding Internasional - THE WORST ANCOVA MODEL.pdf - Published Version
Download (3MB)

Abstract

Data analysis based on two-way ANCOVA models of a numerical variable Y with categorical factors A and B, and various sets of covariates, can easily be done using a software.They are classified into two types of ANCOVA models, namely ANCOVA models with the interfactor A*B as an independent variable, and the additive ANCOVA models without the interaction. I would consider,a additive N - way ANCOVA model isthe worst model among all general linear models with the same set of variables, in a theoretical sense.In addition,based on abalanced panel data with T N  individual time or firm-time observations, various two-way fixed-effects models (TWFEMs) have been presented in the Journal of Finance(JOF).Forexamples,Atanassove(2013)presents several TWFEMs using over 147 thousand firm-time observations, and Vikrant (2013,p.908) presents several TWFEMS and two three-way FEMs,namely Firm,Year,and Industry*Year Fixed Effects Models using 15,310observation with2,948 firms.In fact,a TWFEM is a special additive Two-way ANCOVA model with A is a factor of the N-firms, and Bisa factor of the T-time points, and various sets of covariates. Thence,a TWFEM is representing T N simple linier regressions or multiple regressions with the same slopes of the covariates, and T Ndifference intercepts.Furthermore, note that each of the T Nhomogeneous regressions contains only a single observation,since the data has only T Nobservations.So I would consider TWFEMs are the worst models.To support this statement this paper presents an empirical results based on a data with 3 4firm-time observation, with special notes and comments. Furthermore, for a comparison,this paper also presents general specific equations of Heterogeneous Regressions, two-way ANCOVA models,and TWFEMs,with selected empirical outputs,using EViews.Refer to Agung (2014a,2011)foradditional examples with special notes and comments on various statistical models Keywords: Anacova model, Set of variable

Item Type: Conference or Workshop Item (Paper)
Additional Information: Statistics Department and Mathematics Department Faculty of Mathematics and Natural Sciences State University of Makassar Indonesia
Uncontrolled Keywords: Anacova model, Set of variable
Subjects: FMIPA
KARYA ILMIAH DOSEN
Universitas Negeri Makassar > KARYA ILMIAH DOSEN
Divisions: KOLEKSI KARYA ILMIAH UPT PERPUSTAKAAN UNM MENURUT FAKULTAS > KARYA ILMIAH DOSEN
KARYA ILMIAH DOSEN
Depositing User: Pustakawan Amaluddin Zaihal
Date Deposited: 24 Dec 2020 12:27
Last Modified: 31 Aug 2021 16:40
URI: http://eprints.unm.ac.id/id/eprint/18808

Actions (login required)

View Item View Item
Test Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.